Accuracy of ventilation feedback devices compared with a reference standard during asynchronous ventilation with continuous chest compressions in a simulated cardiac arrest: a prospective non-randomized cross-over study.
Resuscitation plus · 2026-09-01 · Method & validation
Abstract
AIM: This study aimed to evaluate the accuracy of two commercially available ventilation feedback devices (VFD) in measuring ventilation rate and tidal volume during simulated cardiopulmonary resuscitation (CPR) with continuous chest compressions and asynchronous ventilations in the presence of reversed airflow. METHODS: Prospective, simulation-based, non-randomized cross-over study. Twenty emergency department nurse practitioners performed manual ventilation on an intubated high-fidelity manikin during two 1-minute trials (one per device), targeting 10 breaths/min during continuous chest compressions (100-120/min). Ventilation parameters displayed by EOLifeX® and AccuVent® were compared with measurements from a high-fidelity pneumotachograph (HFRP). Data is presented as mean (standard deviation, min-max range). CI = confidence interval. RESULTS: EOLifeX®: From HFRP flow data, 254 ventilations were identified (rate 12/min [1; 8-14], tidal volume 544 mL [32; 482-604]). All showed reversed airflow with an average of 49 mL [13; 19-72]. VFD displayed values overestimated ventilation rate and underestimated tidal volume (111/min [5; 99-121] and 104 mL [14; 68-240]; biases +99 breaths/min, 95% CI 96.6-101.4, and -440 mL, 95% CI -456.3 to -423.7). AccuVent®: From HFRP flow data, a total of 222 ventilations were identified (rate 12/min [1; 8-13], tidal volume 532 mL [35; 460-614]). All showed reversed airflow with an average of 56 mL [15; 22-78]. VFD displayed values were similar for rate (12/min [1; 8-13]) but underestimated tidal volume (360 mL [24; 292-398]; bias -172 mL, 95% CI -191.9 to -152.1). CONCLUSION: During simulated CPR with continuous compressions and asynchronous ventilations, VFDs demonstrated clinically relevant inaccuracies. Validation under realistic resuscitation conditions are needed.